The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts preflight.py and create_feishu_docs.py use subprocess.run to execute system commands. This is primarily used to discover the system PATH by invoking shells (e.g., /bin/bash -lic) and to interact with the lark-cli tool. These operations are restricted to environment configuration and the skill's primary purpose of document management.
[EXTERNAL_DOWNLOADS]: The skill's documentation guides the user to install the @larksuite/cli package via npx. As this is an official tool from a well-known service provider (Lark/ByteDance), this is considered a safe and standard procedure for extending the agent's capabilities to interact with Feishu.
[DATA_EXFILTRATION]: While the skill interacts with a remote service (Feishu), it does so using the user's own authenticated CLI session. The data being sent consists of the mock meeting transcripts provided within the skill's own assets. There is no evidence of sensitive local data (such as SSH keys or environment secrets) being accessed or exfiltrated.

fde-mock-meeting-transcripts